Using Fourier transform near infrared spectroscopy to estimate the nutritional value in whole and milled naked oats

Naked oat is an important grain cereal that originated in, and is cultivated mostly in, China. Naked oat grain has a good processing performance because it is hulless. Its products are important sources of good nutritional protein, dietary fibre, unsaturated fatty acid and polyphenol, all of which have demonstrable benefits for human health. To select the appropriate industry material and develop genotypes with a high nutrient content more efficiently, a fast screening method is needed. In the present study, calibration models were developed from Fourier transform near infrared (FT-NIR) spectra of whole grain and milled grain samples for the estimation of protein, starch, lipid, beta-glucan and avenanthramides content in naked oats from an initial set. The optimised models of protein, starch, lipid, beta-glucan and avenanthramides in whole grain/milled samples indicated coefficients of determination (R-2) of 0.98/0.98, 0.97/0.98, 0.97/0.97, 0.98/0.98 and 0.90/0.90 and root mean square errors of cross-validation of 0.40/0.35%, 0.59/0.52%, 0.27/0.24%, 0.09/0.06% and 60.4/59.2 mg kg(-1). Test validations yielded root mean square errors of prediction of 0.46/0.40%, 0.68/0.63%, 0.26/0.26%, 0.10/0.09% and 47.7/46.3 mg kg(-1) with residual predictive deviations of 5.38/6.30, 5.07/5.15, 6.06/6.24, 5.13/6.17 and 3.37/3.56. These findings showed that FT-NIR could be used as a rapid and accurate method for selecting higher-quality naked oats in a breeding programme as well as for quality control by the food industry.